The present invention relates to novel membrane translocation peptide carrier moieties and membrane translocation vectors comprising a novel peptide carrier moiety together with a cargo moiety, of use in the improved delivery of therapeutic agents into target cells.
The pharmaceutical industry has for many years concerned itself with the efficient delivery of therapeutic agents. This problem may be attributed to the short clearance time of the agent in the body (short half-life), the location of the site of action or possibly the nature of the therapeutic agent itself, for example, its solubility, hydrophobicity etc. Thus, many developments and strategies have been adopted, including formulating the therapeutic agent so as to protect it from a hostile environment on route to its site of action, by for example, enterically coated tablets, controlled release devices and the like.
The development of peptide derived therapeutic agents has posed a further problem due their susceptibility to enzymatic degradation not only in the GI tract but also in the bloodstream. An example of how this problem has been addressed relates to the incorporation of the peptides into liposomes or polymeric microspheres that target the peptides to the lymph system.
A further related problem, especially for therapeutic agents that function intracellularly is the barrier posed by the cell membrane. Thus, it may be possible to increase the half life of the agent or ensure that it passes through the body without being degraded, but many agents must actually enter cells to exert their therapeutic effect.
Homeoproteins are trans-activating factors involved in multiple morphological processes. They bind to DNA through a sequence of 60 amino acid residues, the so-called homeodomain. The structure of this domain consists of three α-helices, interrupted by a β-turn between helices 2 and 3 (Gehring, W. J. et al., (1990) Trends Genet 6, 323-9). The phylogenetic relationship between numerous homeoproteins is striking at the level of the homeodomain and particularly within the third α-helix. This helix is responsible for both the interaction with DNA, as well as the capacity of homeoproteins to translocate across cell membranes to cell nuclei in a non-specific manner.
European Patent 485578 discloses that the homeodomain and specifically, helix 3 of a homeobox peptide, particularly that derived from the Drosophila Antennapedia, is of use as an intracellular transport vector. The patent disclosed that a specific 57 amino acid sequence of a Drosophila Antennapedia homeopeptide (referred to as the pAntp peptide) was capable of penetrating fibroblasts and embryo cells (in vivo). Emphasis was placed upon the last 27 amino acids of the sequence that correspond with the helix 3 and 4. There is no description of the pAntp peptide being linked to any other peptide or therapeutic agent.
Subsequent disclosures (Derossi D et al., J Biol Chem (1994) 269, 10444-10450, Derossi D et al., J Biol Chem (1996) 271, 18188-18193, Joliot A H et al., (1991) The New Biol 3, 1121-1134 and PNAS (1991) 88, 1864-1868, Perez F et al., J Cell Sci (1992) 102, 712-722), have focused on a 16 amino acid synthetic peptide derived from the third helix of the Antennapedia homeodomain that may be used for the intracellular delivery of bioactive products and antisense oligonucleotides. The amino acid sequence of this peptide is RQIKIWFQNRRMKWKK (SEQ ID No. 1) also known as penetratin. In the course of their investigations the above authors synthesized several variants on this sequence, these corresponding to residues 41-60, 41-55 and 46-60 of the pAntp peptide and showed that in all cases, the only peptides to internalise into the cells were those that included the residues 43-58 (Derossi D et al., supra.).
In an effort to prevent the enzymatic cleavage of this peptide Brugidou J et al., (Biochem Biophys Res Comm (1995) 214(2), 685-693) prepared a retro-inverso form (D amino acids in reverse order) of SEQ ID No. 1, substituting the two isoleucine resides at positions 3 and 5 of penetratin with valine and adding a glycine residue at the C-terminus to facilitate binding to a resin. A further retro-inverso form was prepared replacing the extra glycine with a cholesterol moiety attached via a sulfhydryl linker group. The addition of the cholesterol moiety improved penetration due to the increased hydrophobicity of the molecule.
This development of the retro-inverso form of penetratin has given rise to WO 97/12912 that discloses peptides of 16 amino acids comprising between 6 and 10 hydrophobic amino acids wherein the sixth amino acid from either end must be tryptophan. This disclosure attempts to define the minimal characteristics of sequences capable of acting as internalisation vectors as being the retention of a tryptophan residue at the sixth position from the amino terminus and that the peptide contains from 6 to 10 hydrophobic amino acid residues (the classification of hydrophobic amino residues in WO97/12912 is not believed to be in agreement with the generally accepted classification).
From the disclosures discussed above, as summarised in WO97/12912, it has been concluded that essential to the membrane translocating properties of the homeodomain peptides, is the presence of a tryptophan residue as the sixth residue from the amino terminus. Conforming to these requirements has been a penetratin variant of the formula (KWKK)4 (SEQ ID No. 64) which has been described as having translocating ability (Maruta H et al. Cytoskeletal tumour suppressors that block oncogenic RAS signalling. Presented at Anti-Cancer Proteins and Drugs: Structure, Function and Design; 6-9 Nov. 1998, New York Academy of Sciences. Poster/abstract No. 11) and Plank C et, al. (Human Gene Therapy, (1998) 10, 319-332) that discloses a number of branched membrane translocating peptides such as (KWKK)2KGGC, (SEQ ID No. 65), wherein each KWKK (SEQ ID No. 66) is joined to the following lysine residue.